Evidence On Categories And Categorisation

Human knowledge consists of everything that we know. In any attempt to characterise this knowledge, a starting point for research is hard to find. As we have seen in Chapter 9, a distinction has been made between "objects" (dog, cat, dishwasher, spigot) and the "relations" between things (above, below, kick, hit); it is the former we will concentrate on here. Research on object concepts has been heavily influenced by philosophy; especially the British empiricist philosophers (e.g., Locke, 1690) who viewed concepts as being atomic units that were combined in molecule-like ways into more complex structures. We will see that this is a common thread running through research in this area.

In this section, we will review some of the main findings in categorisation before considering the various theoretical accounts of these findings from four positions: the defining-attribute, prototype, exemplar, and explanation-based views. It is only relatively recently that the multiple functions of categories have been considered to be important, and in line with this trend we review the use of categories from multiple perspectives of usage (Sloman & Rips, 1998). Empirical research on categorisation examines the ways in which concepts are used; for example, in making category judgements, making predictions, or explaining differing perspectives on a category. Category judgements can be made about whether a particular instance is a member of the category; for instance, whether the mongrel next door and the winner of Crufts are both instances of the category dog (n.b., we use the convention of writing dog in italics to indicate that we are discussing the concept dog and not the word "dog"). Category judgements can also be used to determine the hierarchical relationship between concepts; for example, how the concepts dog and cat are subordinates (i.e., more specific versions) of the more general animal concept (called the superordinate). Traditionally, these two areas have been heavily researched in the categorisation literature. More recently, the empirical attention has shifted to how concepts are used to predict things; for example, how knowing that someone is called Peter is not that informative, but knowing that Peter is a goldfish allows you to predict that he is likely to swim and eat fish food. The fourth and final empirical area we shall examine is research on the instability of concepts, on how categorisation changes under the influence of differing goals and perspectives.

Category judgements of membership

Intuitively, one of the basic ways in which we use our concepts is in judging whether something is a specific instance of a category. For example, determining whether the animal running across the lawn is a dog or a deer. As such, a large body of work has concerned itself with judgements of category membership. The view taken in object-concept research has been that concepts are defined by attributes; for example, a specific dog is categorised as a dog by virtue of having four-legs, fur, barking, and panting-a-lot. In response to behaviourist accounts of categorisation, some of the earliest cognitive work tried to show that category judgements were rule-governed based on the consideration of such attributes.

Category judgements can be rule-governed

In particular, the influential work of Bruner, Goodnow, and Austin (1956) looked at how people acquire concepts of shapes involving different attributes. In Bruner et al.'s experiments subjects were shown an array of stimuli (see Figure 10.1) that had different attributes (e.g., shape, number of shapes, shading of

Stop Anxiety Attacks

Stop Anxiety Attacks

Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!

Get My Free Ebook


Post a comment